Target-Oriented Single Domain Generalization

Marzi Heidari; Yuhong Guo

REVIEW 3 major objections 5 minor 45 references

Target-Oriented Single Domain Generalization

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper introduces Target-Oriented Single Domain Generalization, in which a textual description of the deployment domain replaces unavailable target images, and shows that a lightweight spectral alignment module (STAR) consistently outpe

desk verdict A promising new problem setting and a clean method, but the paper never proves the text description is what drives the gains—the missing random-text control is the load-bearing gap. read the letter →

arxiv 2509.00351 v1 pith:D3T5CSQQ submitted 2025-08-30 cs.CV cs.AIcs.LG

Marzi Heidari , Yuhong Guo This is my paper · ORCID

classification cs.CVcs.AIcs.LG

keywords singledomaingeneralizationvision-languagemodelsCLIPtext-guidedfeaturealignmentspectralprojectionshiftobjectdetection

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that a one-line natural-language description of the target deployment environment—no target images—can materially improve single-domain generalization. It proposes STAR, which uses a CLIP text embedding as a semantic anchor to re-center source image features and project them onto a target-aligned low-rank subspace. The method is tested on image classification (PACS, DomainNet) and object detection (Diverse-Weather), where it reports consistent gains over prior SDG baselines. If correct, this establishes textual metadata as a cheap, practical substitute for target data under severe distribution shifts.

What carries the argument

The Spectral Target Orientation (STO) operator is the load-bearing mechanism: given a target text embedding φt from CLIP, it re-centers source features as Ω = (z − μz) + φt, computes the SVD of Ω, retains the top k = η·d singular directions to form a projection matrix P_t = V_k V_k^T, projects the original features, renormalizes by norm, and blends with the original representation via α. This is supplemented by Vision-Language Distillation (regressing backbone features to CLIP image embeddings) and Feature-Space Mixup (interpolating between source and target-oriented features).

What would settle it

Train STAR on PACS with a deliberately wrong target description (e.g., 'sketch' when the target is 'art painting'). If accuracy stays at the same level as with the correct description, the gain is not coming from target semantics. Alternatively, set the distillation weight to zero and the blend weight α to 1: if STO alone still produces most of the gain, the target anchor is operating without cross-modal alignment.

Watch

Extended reading notes

Core claim

The central claim is that in the single-domain generalization setting, where only one source domain is available and target images are off-limits, a textual description of the target domain can be converted into a feature-space anchor that guides generalization. STAR operationalizes this by first encoding the text with a frozen CLIP text encoder to obtain a target embedding φt, then translating each batch of source features so its mean coincides with φt, performing SVD on the re-centered features, and projecting onto the top-k singular directions. The target-oriented features are blended with the original ones, a vision–language distillation loss regresses backbone features toward CLIP image

Load-bearing premise

The CLIP text embedding and the backbone image-feature space are aligned well enough that adding the text vector to image features is a meaningful semantic operation rather than an arbitrary shift.

Editorial extensions

If this is right

On image classification, STAR reports average accuracy gains over the strongest prior method: 60.3% vs 57.1% on PACS and 30.0% vs 27.1% on DomainNet, with the largest per-domain gain on Sketch (4.4 points).
On object detection, STAR improves mAP across all Diverse-Weather target domains: Night Rainy mAP rises from 18.7 to 21.0 and Dusk Rainy from 32.3 to 35.2, while also improving source-domain Day Clear detection.
The approach works with different vision–language models: CLIP, BLIP, and LLaVA all yield competitive results, with CLIP best at 59.8% average on PACS.
Ablations show each component contributes: removing the distillation loss costs 4 points, removing the spectral projection drops accuracy to 51.1%, and replacing top-k with bottom-k projection drops to 54.9% versus 60.3% for the full model.
Only a domain-name-level text description is required—e.g., 'Art Painting' or 'Night Rainy'—so the method is applicable whenever practitioners can articulate the deployment environment in words.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The mechanism's success hinges on the backbone feature space and CLIP text space being approximately aligned; a direct test would be to measure whether backbone features of a domain shift closer to the CLIP embedding of the correct target description during training.
If the target description is imprecise or wrong, the anchor could pull features toward an incorrect region; a natural extension is to supply multiple candidate descriptions and let the model weight or select them.
The paper uses the same text description for all classes; richer per-class descriptions (e.g., 'art painting of a dog' instead of 'art painting') might transfer better to fine-grained shifts.
The batch-level recentering is a training-time operation; a test-time variant using target batch statistics would be a direct extension, but it would relax the single-domain generalization constraint that the paper maintains.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Punchline: this is a sensible new problem setting with a clean method and consistent—if modest—gains, but the paper never proves the text description is what's doing the work. A random-text control is missing, and that is the load-bearing gap.

What's new and good: TO-SDG is a practical twist on SDG—you don't have target images, but you can say where you're deploying. STAR is a simple combination of recentering around a text embedding, top-k SVD projection, CLIP distillation, and feature-space mixup; the combination hasn't been published before. The ablations are well structured: the bottom-k projection control is a nice sanity check, and the multi-VLM comparison plus object detection results go beyond what most SDG papers report. The math is elementary and correct.

Soft spots, in order. First, and most important, there is no control where phi_t is replaced with a random vector or an unrelated text phrase. Since the baselines don't get any text at all, the reported gains could come from the distillation and mixup regularizers rather than from the semantics of the target description. Table 5 deepens this worry: three very different text encoders land within 0.8 points of each other (59.0, 59.0, 59.8). A random-text condition would be the direct test, and it's absent. Second, the comparisons against non-text baselines conflate resource and method; a text-aware baseline (e.g., a constant feature offset or a text-prompted head) would be needed. Third, reporting issues: detection tables lack standard deviations despite the five-seed claim, and Table 5's full-model average (59.8) doesn't match Tables 1/6 (60.3). Minor but sloppy. Fourth, the mechanism analysis is asserted rather than shown. Note: the stress-test concern about Eq. (2) being a classifier-absorbable constant shift doesn't quite hold—phi_t is not added to the features that reach the classifier; it shapes the projection matrix via the SVD. But the underlying question of whether the rank-1 update direction is semantically meaningful remains open.

Bottom line: this deserves a serious referee. The setting is worth having, and the results are plausible enough to warrant follow-up. I'd recommend a conditional accept: require a random-text control and a text-aware baseline before the central claim can be accepted.

Referee Report

3 major / 5 minor

Summary. The paper introduces Target-Oriented Single Domain Generalization (TO-SDG), a protocol in which a single source domain is supplemented only by a textual description of the unseen target domain. The proposed method, STAR, injects the target text embedding into source features through three components: Spectral Target Orientation (STO), which re-centers features around the text anchor and projects onto a top-k SVD subspace; Vision-Language Distillation (VLD), which regresses backbone features to CLIP image embeddings; and Feature-Space Mixup (FSM), which interpolates between original and target-oriented features. Experiments on PACS, DomainNet, and Diverse-Weather report consistent improvements over prior SDG baselines in both classification and object detection. The core claim is that textual target metadata alone, without target images, materially improves single-source generalization.

Significance. If the claimed effect is real, TO-SDG is a useful problem formulation and STAR is a practical instantiation: it uses only a frozen VLM and adds a lightweight feature transformation, with no target images. The paper's internal ablations isolate the contributions of VLD, projection, and Mixup, and the authors report five-seed classification results with standard deviations. These are strengths. However, the central scientific claim—that the semantic content of the target text, not merely the presence of an auxiliary vector or of a CLIP-based regularizer, drives the gains—is not directly tested. The main comparisons are against baselines that do not receive the textual description, so the headline margins conflate the extra resource with the method. The paper would be considerably stronger with a no-text / random-text control and with detection variance reporting.

major comments (3)

[§3.2, Eq. (2), Table 5] The central claim is that the textual description of the target domain is what improves generalization. Yet no experiment replaces the text embedding φ_t with a random vector, a constant vector, or a mismatched text string. Because Eq. (2) amounts to adding a batch-wise constant vector, and a linear classifier can absorb a constant offset in its bias, it is possible that much of the gain comes from the spectral/SVD regularization and VLD alone, independent of the text's semantic content. Table 5 shows that BLIP, LLaVA, and CLIP produce nearly identical averages (59.0, 59.0, 59.8), which is consistent with the content of φ_t being nearly irrelevant. A random-text or no-text ablation is necessary to attribute the reported gains to the proposed problem setup.
[Tables 3, 4, 7, 8, 9] Appendix B states that all experiments were replicated five times and results are reported with standard deviations, but the object detection tables report no variance. For a paper whose headline detection claims include a 2.3 mAP gain on Night Rainy (21.0 vs 18.7), the absence of standard deviations (or confidence intervals) in the detection experiments makes it impossible to assess whether the margins are meaningful. The classification tables include std, but the detection tables are the only evidence for the detection part of the contribution.
[§4.2, Tables 1 and 2] The classification baselines (MixUp, CutOut, ADA, ME-ADA, AugMix, RandAug, ACVC, L2D, PR-C, AdvST) do not receive any textual description of the target. STAR receives the target text and uses a CLIP backbone/distillation, so the comparison bundles the extra input modality and the use of a VLM with the proposed algorithm. A more direct comparison would include a version of STAR with a randomly assigned or constant text, or a CLIP-feature-based baseline that does not use target text. Without such controls, the claim of 'superiority' over prior SDG methods is not cleanly attributable to the TO-SDG text signal.

minor comments (5)

[§4.3.2, Table 6] The text refers to 'Lcl' when describing the removal of the classification objective, while the table column is labeled '−w/o Lsup'. Please standardize the notation.
[§4.3.2, Table 6] The label '−w bottom k' is ambiguous; it should read '−w/o bottom-k projection' or similar to avoid confusion with the top-k projection used in the method.
[§3.2.1, Eq. (5)] The rescaling step in Eq. (5) is said to preserve the norm of each batch, but the notation uses ∥Z∥2 for a matrix. Clarify whether this is a Frobenius norm or a per-sample vector norm.
[§4.2, Table 4] Per-class tables are useful, but the main text repeats the same qualitative claims for Tables 7–9. Consider moving some per-class discussion to the appendix, since the main text already includes the key mAP comparisons.
[Appendix B] The detection training uses only 1000 iterations, which seems very small for Faster R-CNN on Diverse-Weather. Please justify this choice and state the batch size, since a short training schedule may affect the reliability of the reported detection numbers.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation; target text embedding is an external input, not a fitted or self-referential quantity.

full rationale

The paper's derivation chain is not circular. The target embedding phi_t = M(T) (Eq. 1) is a frozen CLIP text embedding computed once from the free-form target description and is never fitted to target data or to the method's own outputs. The STO transform (Eqs. 2-7), VLD loss (Eq. 10), and FSM loss (Eqs. 11-12) all take phi_t as an external conditioning input, and the supervised loss (Eq. 9) is evaluated on source labels only. No fitted parameter is renamed as a prediction: the reported gains are measured on held-out target domains that provide no gradient signal during training. The paper contains no load-bearing self-citations (the authors' prior work is not invoked), no imported uniqueness theorem, and no result that is defined in terms of the quantity it claims to predict. One legitimate experimental concern is that the ablations in Table 6 do not include a random-text or constant-embedding control, so the causal role of the text content, as opposed to the structural regularization components, is not isolated; however, a missing control is an evidence gap, not a circularity by construction. The central empirical claim therefore has independent content.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method rests on four hand-chosen hyperparameters (alpha, eta, lambda_mix, lambda_distill) and one unspecified Beta shape; none are derived. The key domain assumptions are the alignment of ResNet-18 feature space with CLIP text space (Eq 2), and the claim that top-k SVD directions of the re-centered source batch are target-relevant. No invented entities are introduced; STAR composes linear algebra operations and losses on existing features.

free parameters (5)

alpha (STO blending coefficient) = 0.9
Controls blend of original and target-oriented features in Eq (7); set by hand, no sensitivity analysis or validation protocol reported.
eta (spectral truncation ratio) = 0.5
Fraction of top singular directions retained in Eq (4); no sensitivity analysis.
lambda_mix (mixup weight) = 0.5
Weight of feature-space mixup loss in Eq (13); chosen by hand.
lambda_distill (distillation weight) = 0.01
Weight of CLIP distillation loss in Eq (13); small value assumed sufficient to align backbone and CLIP spaces.
Beta distribution parameter for mixup
Eq (11) says beta is sampled from a Beta distribution but the shape parameter is never specified; an unstated free choice.

assumptions (4)

domain assumption Backbone features and CLIP text embeddings occupy an approximately aligned coordinate space, so the vector addition in Eq (2) is semantically meaningful.
Invoked in Section 3.2.1; the ResNet-18 backbone is not a CLIP encoder and alignment rests on the small-weight distillation loss (Eq 10, Eq 13).
domain assumption The top-k singular vectors of the source batch re-centered at phi_t (Eq 3) span a 'target-aligned subspace' whose retention improves generalization.
Section 3.2.1, Eqs (3)-(4); target information enters only through the constant offset phi_t, otherwise the subspace is a source-data PCA; no derivation ties these directions to target cues.
domain assumption The textual description (a domain name such as 'Art Painting') is an informative and sufficient summary of the shift to the target domain.
Used throughout the experiments; in all benchmarks the textual description is just the target domain name.
domain assumption The frozen CLIP image encoder provides a semantic geometry worth regressing to (Eq 10).
The distillation objective assumes CLIP's image embeddings are better suited to the target than the backbone's own features; supported only indirectly by ablations.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Target-Oriented Single Domain Generalization." pith.science (2026). https://pith.science/paper/D3T5CSQQ

@misc{pith2026250900351,
  author       = {Pith},
  title        = {Pith review of: Target-Oriented Single Domain Generalization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/D3T5CSQQ}},
  note         = {Machine review of arXiv:2509.00351}
}

read the original abstract

Deep models trained on a single source domain often fail catastrophically under distribution shifts, a critical challenge in Single Domain Generalization (SDG). While existing methods focus on augmenting source data or learning invariant features, they neglect a readily available resource: textual descriptions of the target deployment environment. We propose Target-Oriented Single Domain Generalization (TO-SDG), a novel problem setup that leverages the textual description of the target domain, without requiring any target data, to guide model generalization. To address TO-SDG, we introduce Spectral TARget Alignment (STAR), a lightweight module that injects target semantics into source features by exploiting visual-language models (VLMs) such as CLIP. STAR uses a target-anchored subspace derived from the text embedding of the target description to recenter image features toward the deployment domain, then utilizes spectral projection to retain directions aligned with target cues while discarding source-specific noise. Moreover, we use a vision-language distillation to align backbone features with VLM's semantic geometry. STAR further employs feature-space Mixup to ensure smooth transitions between source and target-oriented representations. Experiments across various image classification and object detection benchmarks demonstrate STAR's superiority. This work establishes that minimal textual metadata, which is a practical and often overlooked resource, significantly enhances generalization under severe data constraints, opening new avenues for deploying robust models in target environments with unseen data.

Figures

Figures reproduced from arXiv: 2509.00351 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 42 canonical work pages

[1]

Generalizing to unseen domains via adversarial data augmentation,

R. V olpi, H. Namkoong, O. Sener, J. C. Duchi, V . Murino, and S. Savarese, “Generalizing to unseen domains via adversarial data augmentation,” in Advances in Neural Information Processing Systems (NeurIPS), 2018

work page 2018
[2]

Learning to learn single domain generalization,

F. Qiao, L. Zhao, and X. Peng, “Learning to learn single domain generalization,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020

work page 2020
[3]

In search of lost domain generalization,

I. Gulrajani and D. Lopez-Paz, “In search of lost domain generalization,” in International Conference on Learning Representations (ICLR), 2020

work page 2020
[4]

Learning transferable visual models from natural language supervi- sion,

A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, et al., “Learning transferable visual models from natural language supervi- sion,” in International conference on machine learning (ICML), 2021

work page 2021
[5]

Scaling up visual and vision-language representation learning with noisy text supervision,

C. Jia, Y . Yang, Y . Xia, Y .-T. Chen, Z. Parekh, H. Pham, Q. Le, Y .-H. Sung, Z. Li, and T. Duerig, “Scaling up visual and vision-language representation learning with noisy text supervision,” in International conference on machine learning (ICML), 2021

work page 2021
[6]

Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,

J. Li, D. Li, C. Xiong, and S. Hoi, “Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,” in International conference on machine learning (ICML), 2022

work page 2022
[7]

Learning to prompt for vision-language models,

K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Learning to prompt for vision-language models,” International Journal of Computer Vision (IJCV), 2022

2022
[8]

Conditional prompt learning for vision-language mod- els,

K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Conditional prompt learning for vision-language mod- els,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), 2022

work page 2022

Show all 45 references

[9]

Clip-adapter: Better vision-language models with feature adapters,

P. Gao, S. Geng, R. Zhang, T. Ma, R. Fang, Y . Zhang, H. Li, and Y . Qiao, “Clip-adapter: Better vision-language models with feature adapters,” International Journal of Computer Vision (IJCV), 2024

2024
[10]

Tip-adapter: Training- free adaption of clip for few-shot classification,

R. Zhang, W. Zhang, R. Fang, P. Gao, K. Li, J. Dai, Y . Qiao, and H. Li, “Tip-adapter: Training- free adaption of clip for few-shot classification,” inEuropean conference on computer vision (ECCV), Springer, 2022

2022
[11]

Improved regularization of convolutional neural networks with cutout,

T. DeVries and G. W. Taylor, “Improved regularization of convolutional neural networks with cutout,” arXiv preprint arXiv:1708.04552, 2017

2017 arXiv
[12]

Augmix: A simple data processing method to improve robustness and uncertainty,

D. Hendrycks, N. Mu, E. D. Cubuk, B. Zoph, J. Gilmer, and B. Lakshminarayanan, “Augmix: A simple data processing method to improve robustness and uncertainty,” in International Conference on Learning Representations (ICLR), 2019

2019
[13]

Randaugment: Practical automated data augmentation with a reduced search space,

E. D. Cubuk, B. Zoph, J. Shlens, and Q. V . Le, “Randaugment: Practical automated data augmentation with a reduced search space,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), June 2020. 10

2020
[14]

Geometry-aware data augmentation for monocular 3d object detection,

Q. Lian, B. Ye, R. Xu, W. Yao, and T. Zhang, “Geometry-aware data augmentation for monocular 3d object detection,” arXiv preprint arXiv:2104.05858, 2021

2021 arXiv
[15]

Attention consistency on visual corruptions for single-source domain generalization,

I. Cugu, M. Mancini, Y . Chen, and Z. Akata, “Attention consistency on visual corruptions for single-source domain generalization,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2022

2022
[16]

Maximum-entropy adversarial data augmentation for improved generalization and robustness,

L. Zhao, T. Liu, X. Peng, and D. Metaxas, “Maximum-entropy adversarial data augmentation for improved generalization and robustness,” in Advances in Neural Information Processing Systems (NeurIPS), 2020

2020
[17]

Adversarial autoaugment,

X. Zhang, Q. Wang, J. Zhang, and Z. Zhong, “Adversarial autoaugment,” in International Conference on Learning Representations (ICLR), 2019

2019
[18]

Adversarial style augmentation for domain generalization,

Y . Zhang, B. Deng, R. Li, K. Jia, and L. Zhang, “Adversarial style augmentation for domain generalization,” 2023

2023
[19]

Learning to diversify for single domain generalization,

Z. Wang, Y . Luo, R. Qiu, Z. Huang, and M. Baktashmotlagh, “Learning to diversify for single domain generalization,” inProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021

2021
[20]

Progressive domain expansion network for single domain generalization,

L. Li, K. Gao, J. Cao, Z. Huang, Y . Weng, X. Mi, Z. Yu, X. Li, and B. Xia, “Progressive domain expansion network for single domain generalization,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021

2021
[21]

Advst: Revisiting data augmentations for single domain generalization,

G. Zheng, M. Huai, and A. Zhang, “Advst: Revisiting data augmentations for single domain generalization,” in Proceedings of the AAAI conference on artificial intelligence (AAAI), 2024

2024
[22]

Faster r-cnn: Towards real-time object detection with region proposal networks,

S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object detection with region proposal networks,”Advances in neural information processing systems (NeurIPS), 2015

2015
[23]

Iterative normalization: Beyond standardization towards efficient whitening,

L. Huang, Y . Zhou, F. Zhu, L. Liu, and L. Shao, “Iterative normalization: Beyond standardization towards efficient whitening,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), 2019

2019
[24]

Two at once: Enhancing learning and generalization capacities via ibn-net,

X. Pan, P. Luo, J. Shi, and X. Tang, “Two at once: Enhancing learning and generalization capacities via ibn-net,” in Proceedings of the european conference on computer vision (ECCV), 2018

2018
[25]

Switchable whitening for deep representation learning,

X. Pan, X. Zhan, J. Shi, X. Tang, and P. Luo, “Switchable whitening for deep representation learning,” in Proceedings of the IEEE/CVF international conference on computer vision (ICCV), 2019

2019
[26]

Robustnet: Improving domain generalization in urban-scene segmentation via instance selective whitening,

S. Choi, S. Jung, H. Yun, J. T. Kim, S. Kim, and J. Choo, “Robustnet: Improving domain generalization in urban-scene segmentation via instance selective whitening,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), 2021

2021
[27]

Single-domain generalized object detection in urban scene via cyclic- disentangled self-distillation,

A. Wu and C. Deng, “Single-domain generalized object detection in urban scene via cyclic- disentangled self-distillation,” in Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition (CVPR), 2022

2022
[28]

Clip the gap: A single domain generalization approach for object detection,

V . Vidit, M. Engilberge, and M. Salzmann, “Clip the gap: A single domain generalization approach for object detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), 2023

2023
[29]

Exploring the limits of out-of-distribution detection,

S. Fort, J. Ren, and B. Lakshminarayanan, “Exploring the limits of out-of-distribution detection,” Advances in neural information processing systems (NeurIPS), 2021

2021
[30]

Towards unified and effective domain generalization,

Y . Zhang, K. Gong, X. Ding, K. Zhang, F. Lv, K. Keutzer, and X. Yue, “Towards unified and effective domain generalization,”arXiv preprint arXiv:2310.10008, 2023

2023 arXiv
[31]

Clipood: Generalizing clip to out-of-distributions,

Y . Shu, X. Guo, J. Wu, X. Wang, J. Wang, and M. Long, “Clipood: Generalizing clip to out-of-distributions,” in International Conference on Machine Learning (ICML), PMLR, 2023. 11

2023
[32]

Leveraging vision-language models for improving domain generalization in image classification,

S. Addepalli, A. R. Asokan, L. Sharma, and R. V . Babu, “Leveraging vision-language models for improving domain generalization in image classification,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

2024
[33]

Arbitrary style transfer in real-time with adaptive instance nor- malization,

X. Huang and S. Belongie, “Arbitrary style transfer in real-time with adaptive instance nor- malization,” in Proceedings of the IEEE international conference on computer vision (ICCV), 2017

2017
[34]

mixup: Beyond empirical risk minimization,

H. Zhang, M. Cisse, Y . N. Dauphin, and D. Lopez-Paz, “mixup: Beyond empirical risk minimization,” in International Conference on Learning Representations (ICLR), 2018

2018
[35]

Cutmix: Regularization strategy to train strong classifiers with localizable features,

S. Yun, D. Han, S. J. Oh, S. Chun, J. Choe, and Y . Yoo, “Cutmix: Regularization strategy to train strong classifiers with localizable features,” in Proceedings of the IEEE/CVF international conference on computer vision (ICCV), 2019

2019
[36]

Randaugment: Practical automated data augmentation with a reduced search space,

E. D. Cubuk, B. Zoph, J. Shlens, and Q. V . Le, “Randaugment: Practical automated data augmentation with a reduced search space,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2020

2020
[37]

Deeper, broader and artier domain general- ization,

D. Li, Y . Yang, Y .-Z. Song, and T. M. Hospedales, “Deeper, broader and artier domain general- ization,” in IEEE/CVF International Conference on Computer Vision (ICCV), 2017

2017
[38]

Moment matching for multi- source domain adaptation,

X. Peng, Q. Bai, X. Xia, Z. Huang, K. Saenko, and B. Wang, “Moment matching for multi- source domain adaptation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019

2019
[39]

Autoaugment: Learning augmentation policies from data,

E. D. Cubuk, B. Zoph, D. Mane, V . Vasudevan, and Q. V . Le, “Autoaugment: Learning augmentation policies from data,” arXiv preprint arXiv:1805.09501, 2018

2018 arXiv
[40]

Koltchinskii,Oracle inequalities in empirical risk minimization and sparse recovery problems: École D’Été de Probabilités de Saint-Flour XXXVIII-2008

V . Koltchinskii,Oracle inequalities in empirical risk minimization and sparse recovery problems: École D’Été de Probabilités de Saint-Flour XXXVIII-2008. Springer Science & Business Media, 2011

2008
[41]

Unified deep supervised domain adaptation and generalization,

S. Motiian, M. Piccirilli, D. A. Adjeroh, and G. Doretto, “Unified deep supervised domain adaptation and generalization,” in IEEE/CVF International Conference on Computer Vision (ICCV), 2017

2017
[42]

Domain generaliza- tion by solving jigsaw puzzles,

F. M. Carlucci, A. D’Innocente, S. Bucci, B. Caputo, and T. Tommasi, “Domain generaliza- tion by solving jigsaw puzzles,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019

2019
[43]

Addressing model vulnerability to distributional shifts over image transformation sets,

R. V olpi and V . Murino, “Addressing model vulnerability to distributional shifts over image transformation sets,” in IEEE/CVF International Conference on Computer Vision (ICCV), 2019

2019
[44]

Visual instruction tuning,

H. Liu, C. Li, Q. Wu, and Y . J. Lee, “Visual instruction tuning,”Advances in neural information processing systems (NeurIPS), 2023

2023
[45]

Deep residual learning for image recognition,

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 2016. 12 Algorithm 1 Training Algorithm for STAR Input: Source dataset Ds, target text T , M, M′ fθ0 a...

2016

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [1]

Generalizing to unseen domains via adversarial data augmentation,

R. V olpi, H. Namkoong, O. Sener, J. C. Duchi, V . Murino, and S. Savarese, “Generalizing to unseen domains via adversarial data augmentation,” in Advances in Neural Information Processing Systems (NeurIPS), 2018

work page 2018

[2] [2]

Learning to learn single domain generalization,

F. Qiao, L. Zhao, and X. Peng, “Learning to learn single domain generalization,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020

work page 2020

[3] [3]

In search of lost domain generalization,

I. Gulrajani and D. Lopez-Paz, “In search of lost domain generalization,” in International Conference on Learning Representations (ICLR), 2020

work page 2020

[4] [4]

Learning transferable visual models from natural language supervi- sion,

A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, et al., “Learning transferable visual models from natural language supervi- sion,” in International conference on machine learning (ICML), 2021

work page 2021

[5] [5]

Scaling up visual and vision-language representation learning with noisy text supervision,

C. Jia, Y . Yang, Y . Xia, Y .-T. Chen, Z. Parekh, H. Pham, Q. Le, Y .-H. Sung, Z. Li, and T. Duerig, “Scaling up visual and vision-language representation learning with noisy text supervision,” in International conference on machine learning (ICML), 2021

work page 2021

[6] [6]

Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,

J. Li, D. Li, C. Xiong, and S. Hoi, “Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,” in International conference on machine learning (ICML), 2022

work page 2022

[7] [7]

Learning to prompt for vision-language models,

K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Learning to prompt for vision-language models,” International Journal of Computer Vision (IJCV), 2022

2022

[8] [8]

Conditional prompt learning for vision-language mod- els,

K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Conditional prompt learning for vision-language mod- els,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), 2022

work page 2022